Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
BIGl£EDICAL RESEARCH AND DEVELOPMENT: MEASURING THE RETURNS ON INl7ES AT Jeffrey E. Harris Massachusetts Insti~cu~ce of Technology and Massachusetts General Hospital Can we measure ache benefits of research and development (R&l)) in the biomedical sciences? How do *che benefits of basic research compare with those of applied research? How do ache returns on public funding of biomedical R&D compare with re~cu~s.s on in~resemene in the priorate sector? The past four decades have Aid essed an explosion in government and private funding of biomedical research, in drug and instrument patents, in biomedical research manpower, and in new biomedical j ournals . Concurrently, ache morbidity and mortality from a number of diseases have been falling. Yet, we remain far from gauging ache quantitative return to the broad na~cional investment in biomedical research, even farther from gauging the return to federal investments, and still farther from offering clear guidelines for ache allocation of federal biomedical research dollars. This paper poses some knotty conceptual and methodological problems that have not been confronted squarely in past work on the ra turns to biomedical R&D. The main points can be summarized in nine teas ic propos itions: (1) In the biomedical sciences, the separate contributions of basic and applied research in biomedicine can be difficult to distinguish. In many instances, basic and applied research have worked synergistically to yield an important clinical advance. In a few striking cases, highly focused, applied research accidentally has opened new a~res~ues of basic, nondirec~ced inquiry. (2) Similarly, the separate contributions of public and pri~ra~ce investments in biomedical R&D are difficult to distinguish. In many instances, the benefits from public investments have depended on continued research and development in the priorate sector. (3) Not all biomedical innovations have arisen from biomedical research and development. Investments in basic and applied research in the physical. sciences, in computational sciences, and in military applications have produced significant innovations in biomedicine. - 195 -
(4) The relative contributions of domestic and foreign inves~cments in biomedical R&D will become an increasingly impor~can~ issue. In some instances, biomedical innovations originated from foreign governments or 'multinational firms. (5) Assignment of improvements in health deco specific biomedical innovations is not always possible. Many innovations were adopted before controlled studies of their medical benefits were performed. Observed improvements in health may have resulted from public health measures or changes in life style and the environment, rancher than from specific biomedical ir.no~rations. ~ 6 ~ The economic valuation of improvements in health raises important conceptual ques tions . In particular, innovations that prolong life result generally in increased economic transfers from younger, productive generations to older, less productive generations . (7) The valuation of improvements in health likewise presents methodological challenges. Although considerable atten~cion has been devoted to valuing loss of life, the state of the art In gauging improvements in quality of life is far less advanced ( 8 ) Some assessments of the return `~ on biomedical R&D have been based on es timated gains in national output or productivity . Such assessments may understate substantially the public' s willingness to pay for certain innova~cions. Typically, such assessments ignore the profit s of prince firms that have appropriated new Icuowl~edge and techniques . (9) New methods Deco assess the benefits of biomedical inno~rat'ans are needed. A little - explored area of inquiry is the economic returns to innovations that hare been appropria~ced fully by the priorate sector. Many of these concep~cual and methodological. problems apply also to ache assessmen~c of publicly funded R&~) in other areas. This paper will point out ache special ways that such difficulties show up in the biomedical sciences, focusing especially on possible new areas of research. BASIC AND APPLIEI) RESEARCH! In a widely cited study, Comroe and Dripps identified the key scientific papers underlying the top ten clinical advances in cardiovasculax-pulmonary medicine between 194S and 19tS. For all advances combined, 41 percent of the key papers were not clinically oriented to a specific disease at the time of the research. Most of the papers described basic rather than applied research. Only a minority described new apparatus or techniques, synthesized past work, or reported new disease entities. 196 -
It hardly is surprising that basic, nontargeted research has made a substantial ul~cimate contribution to clinical medicine. The more critical - - and less recognized- - result of the Comroe-l~ripps work is the complex synergism between the basic and the applied. ~us, performance of the first successful open heart operation in 1954 required the invention of the artificial heart-lung bypass. We bypass in turn required a pump (heart) and oxygenator ~ lung) ~ A pump that did not damage blood in turn required the development of an anticoagulant ~ the drug, heparin) . Likewise, the des ign of ache oxygenator required an advanced understanding of gas exchange in the lung. Besides the heart- lung bypass, open heart surgery required botch applied advances ire diagnostic techniques (cardiac catheterization) and basic advances in pharmacology ~ anesthetics and neuromuscular blocking agents) and physiology (fluid and electrolyte balance ~ . In the case of poliomyeli~cis, Comroe and Dripps classified as clinically disease-oriented the Nobel- Prize winning research of Enders, Weller, and Robbins on extraneural culture of polio Citrus in vi~cro. But the research was a critical, basic step in the development of all antiviral vaccines O Likewise ~ ache n iron lung" ~ a device deco expand the lungs by creating a negative pressure gradient outside the body) was an applied response to polio- induced respiratory paralysis. Yet, the iron lung stimulated ache basic research underlying the development of positive pressure ventila~cors. In some cases, highly focused, applied research has opened a new line of basic study accidentally. We practical development of neuroleptic drugs led Deco basic research on brain neurotransmitters. More recently, the analysis of con~c~inants of street drugs by clinical toxicologists has led deco the development of a new animal model for Parkinson' s disease. lathe presence of synergistic interac~cions between basic, nondirected research and applied, targeted research complicates any analys is of the returns to Et&D investments . It is conce ivable that only the joint returns from a mix~cure of basic and applied research could ever be identified in empirical work. PUBLIC ANI:) PRIVATE INVESTMENT Typically, economists cite two main reasons for public (as opposed to priorates investments in research and development. First, ache potential gains from the investment may be public goods; that is, the benefi~cs cannot be appropriated exclusively by priorate parties through pater~cs, trade secrets, or firs~c-mover advantages. Second, certain potentially fruitful proj ecus entail far greater costs and far more uncertainty than even the larges~c corporations are prepared to bear. _ 197
These no reasons justify many government- funded R&D proj ects in the biomedical sciences. Thus, a priorate consortium of firms was not likely Deco perform the National Heart, Lung, and Blood Institute ~ s Multiple R:sic Factor Intervention Trial, a f:veoyear, quarter-billion dollar study of the effects of a multifaceted health prevention program on the mortality from coronary heart disease. As many publicly funded projects move forward, however, the initially insurmountable setup costs begin to fall, ache uncertainty becomes more manageable, and certain narrower portions of the pro; ec become appropriable . At some point, privately funded Rhr) becomes feasible; in fact, there may be sustained periods when both public and private funding coexist. In some cases, no doubt, the transition from public to private activity has been smooth; in others, disorderly and wasteful. In still others, public and private funding act synergistically. In the early 1970's, DeLuca, at the University of Wisconsin identified a vitamin D3 metabolita (calcitriol) for correction of abnormal calcium metabolism, especially in patients undergoing renal dialysis. DeLuca applied for patents, which he assigned to the Wisconsin Alumni Research Foundation (WARF)o Because federal funds were used in the WARE research program, the U. S . government retained rights to cancel exclusive licenses issued by WARE. In Burn, WARE granted Hoffman-La Roche, Inc., a license Deco develop practical applica~clons. Therea.feer, Roche developed a practical scheme for synthesis, completed the necessary toxicologic studies, and collaborated with other independent investigators in the design and execution of clinical studies ~ see Faust ~ . The development of calcieriol in ache priva~ce sector has stimulated new pubJ-icly sponsored studies of calcium metabolism in various disease states. In the case of the totally implan~cable artificial heart, public and private activity were mixed in much mole complex ways (see Decries, Carter, L. F. Rothschild et al., used Strauss ~ . By the early 1960's, many private researchers (DeBalcey, Rantrowitz, Kolff, and ethers), often unassisted by public funds, had begun to world on mechanical derricks to aid a failing heart. These researchers (notably DeBakey) pressed Congress for a maj or federal commitment to an artificial heart. By 1965, the Artificial Heart Program ~ a targeted, system=-orien~ced, contract-based scheme of federal. funding) was in place formally at the National Institutes of Health. During ache late 1960 ' ~ and ear] y 1970 ' s, the Artificial Heart Program underwent a number of setbacks and reevaluations. Two major engineering problems--the development of a biocompatible surface material and the design of an implantable power source- - remained unsolved. Despite active intervention by the Atomic Energy Commission, the possibility of a nuciear-powerad heart did not materialize. Concurrent advances in understanding ache prevention and treatment of coronary heart disease made the permanent, total artificial heart ~ ess attractive. Research on the totally - 198 -
ir~plantable heart was nearly abandoned in favor of temporary ventricular assist devices. But the original, federally funded program already had reduced the setup costs and uncertainty inherent in ache technology enough to make the totally implan~cable artificial heart a private sector pro] ect. Moreover, continued federal funding of left ventricular assist devices had maj or spillover effects In the development of brocompatible surfaces, valves, and power sources. By 1976, Willem Wolff had established a separate company to produce the Jack heart. In 1981, the firm sought and obtained priorate venture capital. In 1982, Barney Claric received the first implant. In 1983, a public stock offering was held. By 1985, it become increas ingly apparent that the original biocompa~cibility problem had not been solved: Sub] ects undergoing implants appeared to suffer embolic strokes. Now, the research effort has shifted back to federally funded development of temporary assist devices. lye retrovirus that causes acquired immunadeficiency syndrome (AIDS ~ was discovered in publicly sponsored laboratories in ache Uni ted S tates and France . The discovery led rap idly to the development Of immunoassays to detect antibodies to the virus in blood samples. The patents for such im~unoassays were, in turn, licensed deco priorate firms. The antibody tests, produced privately by public licenses, have yielded benefits to the private sector ~ through screening of donated blood) and the public sector (as an epidemiologic research tool). Publicly funded research in molecular genetics has engendered priorate firms that employ recombinant DNA techniques deco synthesize scarce, important b iologicals in large quantities.. Such private firms have ~ oined government agencies in funding biotechnology research in prince universities (see Blumenthal et al. ). Men one such firm synthesizes human growth hormone, certain patients with hormone deficiencies will benefit. Moreover, ache firm itself makes a profit, while other firms that previously had supplied an impure form of the hormone take losses. These examples illustrate the potential difficulties in dissecting ache retunes of public ant priorate investments in biomedical R&D. they also show how private gains may be relevant to a calculation of the returns on public investment. In Chose cases where public funding generates priorate economic activity, the resulting profits of private firms should be included with consumer benefits In the total return calculation. No empirical. analysis of returns to public fuming of biomedical R&D appears to have ons idered such a pass ibility . - L99
BIOMED T CAL INNOVATIONS AND lIONBIOMEDICAL INVESTMENTS Often, research and development in nonmedical areas of science and engineering have produced unintended benefits for biomedical science. Such snillo~rer effects cannot be ignored in a proper valuation of the returns to biomedical Rho. U1 trasound imaging of internal organs was an outgrowth of sonar, a technique developed originally for military use during World War I and applied to the industrial detection of metal flaws in the 1930 ' s . Although the technique was first used in an attempt deco map The brain in 1943, commercially produced diagr~ostic ultrasound equipment was not available until 1963. Me current surgical treatment for diabetic vascular degeneration of the retina arose through application of laser technology. Direct visualization of internal organs ~ through the esophagoscope, ache flexible sigmoidoscope, the bronchoscope, even the office otoscope was enhanced markedly by the development of fiber optic illumination. Many new, noninvasi~re eechn~q'~es for imaging of inte ~ :~al organs stem from advances in computational science ~ the C scanner), radioisotope and nuclear chemistry ~ galleon and thallium scanning, positron emission tomography), and nuclear magnetic resonance (I scanning). A lit~la-explored question is whether investments in biomedical R&D have yielded significant, unintended benefits in other areas of science and engineering. A case in point is the influence of the neurosciences on the field of artificial intelligence. DOMESTIC VERSUS FOREIGN INVESTMENTS The computed tomographic scanner was first marketed by a foreign corporation. Now, many biotechnology firms are multinational entities. New pha`"aceu~cicals often are developed and tested abroad before introduction into the United S tates . We need to recognize the international scope of biomedical research in any assessment of the returns to U.S. government funding. Thus far, such a possibility has not been addresses seriously. Yet, how are we to gauge the returns to IJ. S . gove~ent- funded research on hen retroviruses when the AIDS virus was discovered concurrently in France? How are we to assess the returns to a program of vaccination against AIDS when the likely site for field testing would be Africa? To date, no study of the benefits and costs of biomedical R&D that frames ache p rote lem In an open economy has been conducted . _ 900
ASSIGHiNG Him BENEFITS TO STINT F I=ES~=S ~ or decades, the biomedical literature has been replete with evaluations of ache effects of various interventions on the incidence and severity of numerous diseases. Although much of the literature has been qualitative, increasing attention has been given deco the quar~titati~re measurement of the costs and benefits of specific preventative and curative measures. The field has grown so rapidly in the past decade as go constitute a separate disciplines edical technology assessment. Chile much of the literature addresses narrowly defined interventions ~ such as surgery for a specific illness ), some ambitious studies have attempted to gauge the net benefits of a broader class of public expenditure programs. Among the seminal studies was that of Klarman, who analyzed the benefits and costs of syphilis control programs . Others in the lace 1960 ~ s produced benefit~cost analyses of chronic renal dialysis. In an influential paper, Weisbrod computed rates of return on in~restmen~c In poliomyelitis research in the range of 4 to 14 percent. More recent work includes the retune on investments in vaccinationl~or measles and rubella ~ see Axnicic et al ~ and Schoenbaum et al . ~ . The most salient characteristic of such rate-of-return studies is that they are confined to unambiguous cases where a specific biomedical innovation resulted in prevention or cure. Unfortunately, for a much wider range of research activities, rhe quantitative link between funds expended and cases prevented or cured is much more difficult to assess precisely. The age - ad; usted death rate from coronary heart disease has been declining for almost two decades. Among the factors most likely to be responsible are the decline in cigarette smoking, chiefly in men; the treatment of hypertension, chiefly in blacks; overall declines in saturated fats in the died ; the advent of acute coronary care; and the availability of new drugs for the treatment of coronary heart disease. Although coronary artery bypass grafting could have had ~ s ignifican~c impac~c on coronary rares, it is difficult to gauge with any precision. After substantial research on the pathology of an acute hear,: attaclc and multiple clinical evaluations of clots dissolving drugs, thro~olytic treatment of heart acacia will undergo widespread adoption without a precise estimate of its benefits . This does not mean that the cause of the decline in coronary mortality is unknown. lathe point is Chat a diverse mix of biomedical investments has indeed paid off. But, the individual components of the investment portfolio cannot be ranked systematically. The case of cancer treatment is similarly problematic. No doubt, important advances in chemotherapy have resulted in improved subdural 2 0 1
and tower mortality from Hodgkin' s disease, lymphatic leukemia, and certain other cancers. If we were to attribute these improvements in survival to the development of a few drugs--vincristine, daunorubicin, adriamycin, etc . - - then the retuners on chemotherapy would be quince impressive . Yet, for the maj or fatal cancers - - lung, breast, and calorec~cai- - the gains from treatment are not as clear (see Bailar and Smirch 3~. Accordingly, the overall returns from the aggregate investment in cancer treatment--including screening of chemotherapeutic agents' development of powerful irradiation devices, clinical trials of radical versus limited surgery, immunotherapy, and bone marrow transplantation--would be considerably lower. One avenue of inquiry is the computation of wide (but possibly still informative) bounds on the returns to various types of research. The upper bounds would come from instances where specific health improvements were attributed to specific innovations. The lower bounds would derive from the diluted returns to a broader class of investments. VALUING HEALTH BENEFITS Quantitative assessment of the benefits of various health improvements has a long history, extending as far back as the origins of public health statistics. In the early decades of the century, assessments of health outcomes were devoted mostly to incidence and mortality rates. In the late 1920's and early L930's, however, public health researchers in Massachusetts already had begun to investigate the impact of cancer and other chronic diseases an employment and income (see Bigelow and Lombard ). The 1950's witnessed an enhanced interest tp attaching economic values t: health outcomes (see Laitip 5 and rein ). In the 1960's, Weisbrod, 7 Mushkin,~8 and Ricet9 formalized a method for valuing health outcomes that has come to be known as She ~cost-~-iliness" approach: (See also2iooper and Rice,224Hodgson, Hodgacn and Meiners, 2 Salke~,er, and Rice et al. ~ In the cost-of- illness approach, economic gains and losses from changes in disease rates have two components: effects on morbidity and morta3-ity, ant effects on health care expenditures. lathe former gauges the changes in productivity and earnings that result from Maceration in health status. The lancer gauges the displacement of earnings to health care (and, thus, away from other uses). Taken together, the two components measure the impact of disease on disposable income. The cost-of-illness approach needs to be applied carefully. Spending on medical care itself generates income. Reduction in disease rates will cause certain resources to be diverted from the health care sector to other uses. If those resources earned more in the health care sector than in their next best use, then the toss in - 202 -
profits due to reduced demand should be subtracted from the other benefits of disease reduction. There are, however, two more serious concerns about ache cos ~ - o f - illness approach. First, the benefits of health improvements are weighted too heavily by their impacts on productivity and earnings. Thus, in principle, the complete cure of a disease that struck only septuagenarians would yield no benefits except for its effects on medical expenditures. Yet, clearly, our society is willing to pay to imp rove and prolong the lives of older nonproductive persons, even if such gains in health do not enhance the gross national product (GNP) . Even during one' s younger, productive years, impacts on productivity and earnings may not capture the effects of health on the quality of life. The second concern arises from the existence of intergenerational transfers' such as those inherent in the Social Security system. In principle, improvements in longevity coul.d increase ache net ~crar~sfers from younger, more productive generations to older, less productive generations . Such intergenerational bans fers could reach the po int where the additional benefit of suppartir~g an older person exceeded ache additional cost of extracting a dollar from someone younger. Considerably more research has been devoted to the first source of bias than the second. Typhus, researchers have proposed a r.nmber or alternatives to lost years of productivity, including years of "potential' life lost, "expected" life lost, "discounted expected" life lost, en: 4 "qualit~6adjusted" expected i:'fe lost (see Zecichauser and Shepa:9, Vaupel, kneeler and Cretin, and Mendeloff ). As Vaupel and Mushkin hairs demonstrated, the relative impacts of different diseases will depend on which measure is used. While cardiovascular disease is the leading cause of death in ache United States, cancer is almost as significant a cause of early death. While degenerative ; oint disease may contribute little to mortality--as gauged by any of the above measures--it can cause maj or deterioration in the quality of life. Research on the valuation of the quality of life is still in its inchoate phase ~ see Saikever 1) One proposed alternative is the "willingness - to In method f or valuing health improvements (see Jones-Lee, Acton, Conley,34 and Arthur ). An improvement in health, at least in theory, is worth what people are willing to pay for it. Me problem is not in theory but in practice: It has been quite difficult to come by natural experiments to measure implied willingness to pay. Heasured wage differentials between hazardous and nonhazardous occupations, for example, are confounded by workers' attitudes coward risk. b£easuremen~c of willingness ~co pay from ac~cual market data requires the assumption tha~c individuals are fully informed and make wholly voluntary choices - - an assumption that often is unwarran~ced. 03
Nevertheless, empirical studies have shown consistently that the willingness- to-pay measure of the value of life exceeds the productivi~y-based measure. This finding should not be surprising. Society is willing to pay more for an individual's health and longevity than that individual can contribute to the GNP. INTERGENERATIONAL TRANSFERS Consider a simple re~ciremen~c plan, in which subscribers pay premiums prior deco retirement, and in which those surviving past retirement receive an annuity as long as they remain alive. Suppose that ache premiums for such a plan are adjusted freely to corer anticipated payments. Under such a simplified plan, improvements in longevity could make premiums rise to pay for the an~cicipated increase in annuity payments . Our social insurance system, however, is not so simple. We have disability payments and death benefits . Premiums do not adj ust automatically. In practice, it is perfectly possible that certain health improvements can decrease the net bans fer from the young and healthy to the old and infirm. The reduction in disability and deal h benefits could outweigh the old age sur~s~rors ' benefits . If a particular health improvement reduced mortality rates in the preretiremene ages, Men longer lived subscribers might pay more into the system well before they extracted their premiums. Accordingly, the effects of improved health and longevity on intergenerational transfers are far from clear. Unfortunately, systematic analyses of the effects of health improvements on the Social Security system and on other foes of social insurance have been lacking. The best we can say is that direction and magni~cude of the intergenerational transfers will depend an the specific health improvement under consideration. Compared to the direct effects on indi~ldual heal th, the welfare effects of such transfers probably would be second ordere PRIVATE GAINS AS A LOWER BOUNI) ON PUBLIC RETURNS Of all the unanswered ques~cions posed thus far, one stands out as particularly imports: Cat proport:~ on of the returns to public in~resment in biomedical R&D are appropriated ultimately by private firms? If the fraction turned out to be substantial, it might at least be possible to place a conservative lower bound on the public return by examining the purely priorate gains. Lee ideal case is where the government invests in R&D and licenses its discoveries deco private firms, which then are permitted to function legally as discriminating monopolists. In that case, public gains equal priorate profits. When exclusion is incomplete or 904
monopoly is not exercised fully, then consumers will gain as well. When an innovation supplants a previous product or technique, the losses of rival firms may have to be sub tracked . The research question is: Are there enough cases of fully appropriated private gains to s tudy in detail? How much of the return Deco publicly funded R&D on neuroleptics was captured ul~cimately by the manufacturers of benzodiazepines? How much of the retune: to research on calcium-channel blockers was captured by pharmaceutical firms? How much of the return to research on infarction reduction was appropriated by the manufacturers of the intra-aortic balloon pup? It would be presumptuous to estimate at this juncture what fraction of public R&D is appropriable ultimately. Nevertheless, i, is much more tractable a problem to measure the returns on prince activity, either by accounting analyses or by studies of stock price movements. In selected instances, ii may be possible §~~ gauge both sac ial and priorate gains ~ see Bailey 6 and Schwartzman ~ . FINAL CO=~S: WINNERS VERSUS LOS ERS 'pith all the complexities raised here, how can we possibly offer even the simplest quad itat~ve guidelines for the allocation of federal funds for ~ iomedical Red)? Two polar strategies for research funding may be distinguished: "playing the losers and "playing the winners n In the play-~}~e-loser view, funds should be allocated to disease areas in proportion to the current burden of the disease, irrespective of prospects for successful research. In the play-the-win~ser view, funds should be allocated to those areas of research that currently are yielding pos Stile gains, irrespective of disease burden. In the play- ache- loser strategy, the probability of a large gain is unknown. In the play- the-winner strategy, there is a large probability of an unknown gain. To take some examples, current research on Alzheimer's disease is closer to a play-the-loser strategy, while current research on AIDS is closer to a play- the-winner strategy. The situation, of course, could change rapidly. There may be a breakthrough in Alzheimer's. Or, AIDS may become so burdensome as to demand ex~censi~re investments even when no breakthrough is at hand. Obviously, the optimal strategy is some mix of these two polar cases. The question is: How do we decide on the mix? Much can be gained by merely attempting to classify potential areas of funding along ache winner- loser continuum. Once a crude classification is fo~mula~ced, the main principle will be 205 _
diversification of strategies for losers and concentration of strategies for winners. Thus, in Alzheimer's research, the portfolio of projects should be diversified, spreading funding over as many different ideas as possible. This means everything from genetic concordance studies to neuroanatomy to clinical trials of multiple agents. In AIDS, the stray tentacles should be pulled in and efforts aimed straight at the prey--whether he be antiviral agents, vaccines, methadone, or condoms . _ 206
REFE^RENCES 1. J. Comrae and R. Dripps. "Scientific Basis for the Support for Biomeedical Science, ~ Science, Vol. 192~1976), pp lOSolll. R. E. Faust. nThe Role of the Pharmaceutical Industry in Collaborative Research: The Calciteriol Storying In Biomedical Innovation. Edited by E. B. Roberts et al. Cambridge, MA: MIT Press, 1981, pp. 144-151. 3. W. C. DeVries. "The Total Artificial Heart Replacement i n Man." A proposal to the Review Committee for Research wi th Human Subj ects ,, University of Utah Medical Centeer, Salt Lake Ci By, February 27, 1981. 4. He N. arch. Are Business Behind Barney Clark's Heart, Money Magazine, To 1 . 12 ( 19 8 3 ~ 9 pp e 130 ~ 144 ~ 5. L. F. Rothschild, Unterberg, Towbine et al. Prospectus. Kolff Medical, Inc. 1, 500, 000 Shares Come-on Stock. Filed with the Securities and Exchange Commission, July 22, 1983. 6. M. J. Strauss. The Political History of the Artificial Heart, New Englar~d Journal of Medicine, Not 310~1984), pp. 332-336. 7 0 D . Blumenthal, M. Cluck, K e S o Louis ~ and 1) . Wise . ~ Indus trial Support of University Research in Biotechnology, ~ Science, Vol . 231 (1986 ), pp . 242- 246 . 80 Office of Technology A.ssessem:nt. Strategies for Medical Technology Assessment. Washington, DC: US. Government Printing Office, 1982. 9. H. E. Klarman. Syphilis Control Programs. n In Measuring cbe Benefits of Government Inves~cmen~s. Edited by R. Dorfm~n. Washington, DC: Brookings Institution, 1965, pp. 367-410. 10. B. A. Weisbrod. "Costs and Benefi~cs of Medical Research: A Case Study of Poliomyelitis, " Journal of Political Economy, Vol. 79 (1971), pp . 527-544. 11. N. U. Airsick, S. M. Shavell, and J. J. Wi~cte. "Benefi~cs Due to Immunization Against Measles," Public Realch Reports, Vol. 90 ~ 1975), pp . 205 - 207 . 12. S. C. Schoenbaum, J. N. Hyde, fir., L. Bartoshes~r, and K. Crampton. ~Benefi~-Cost Analysis of Rubella Vaccination Policy," New England Journal of Medicine, Sol. 294~1976), pp. 306-310. _ 90 7
13. J. C. Bailar III and E. M. Smith. "Progress Against Cancer?" New England Jour`:al of Medicine, Vol. 314~1986), pp. 1226 - 1232. . G. H. Bigelow and H. L. Lombard. Cancer and Ocher Chronic Diseases in Hassachuse~~s. Boston: Houghtos~ Mifflin Co., 1933 . 15. H. Laitin. nThe Economics of Cancer. n Unpubl'shed doc~coral disserta~elon, Harvard University, Cambridge, MA, 1956. . . R. Fein e Economics of Mental Illness . New York: Sas ic Books, 1958. 17. B. A. Weisbrad. Economics of Public Realrb. Phil~delph, a: University of Pennsylvania Press, 1961. 18. S. Muslin. "Health as an Investment, ~ Journal of Political Economy, Clot. 70~1962), pp. 1~29-1-S9. 19 . ~ . P . Rice . Es him tiny she Cos tic o f I] lness . Heal th Economics Series No. 6 Department of Heal th, Education, and Welfare Publication No . 947-6 . Washington, DC: U. S . Go~rernmen~c Prin~cing Office, 1966 . 20. B. S. Cooper and O. P. Rice. "The Economic Cost of Illness Revisited," Social Securic:T Bulletin, trot. 39~1976), p. 21. T. A. Hodgson. "The State of the Art of Cos~c-of- Illness Estimatas," Advances in Health Economics and Health Ser~r~ces Research, Vol. 4~1983~' pp. -129-164. 22. T. A. Hodgson and M. R. Meiners. "Cost-c'£-lilness Methodology: A Guide to Current Practices and Procedures, ~ Milbank Memorial Fund Quarterly, Vol . 60 (1982), pp ~ 429 -462 . 23. 13. SO Salkever. Morbidity Coses: National Estimates Id Economic Determinants. Research Scary Series, National Center for Heal~ch Services Research. Depar~cmen~c of Health and [lumen Services Publication No. (PHS) 86-3393. Washington, OC: U. S . Government Printing Office, 1985. 24. I). P. Rica, T. A. Hodgson, and A. N. Kopstein. "The Economic Cost of Illness: A Replication and Update," Realth Care Financing Retries, Vol. 7~1985), pp. 61-80. 25 . R. Zeckhauser and I) . Shepart e 'here Now for Saving Lives? Law and Contemporary Problem, Voi. 40~1976), pp. 4-45. 26. J. U. Vaupel. "Early Death: An American Tragedy, n Law and Contemporary Problems, Vol. 40~1976), pp. 73-121. 908
. E. B. Keeler and S. Cretin. Miscounting of Lifesaving and Other Nor~mane~cary Effects, n Management Science, Sol . 29~1983) , pp. 300 - 306 . 8 . J . t£endeloff . "Measuring Elus ive Benefits: On the Value of Health, " JOur;`a1 of Ideal oh Pol i tics, Pol icy and Law, ~1~1 . 8 (1983), pp . :54-580 . 29. 11aupel, op. cit. 30 . S . Hushkir~. Biomedical Research: Gosts and Benef its . Cambridge, MA: Ballinger Publishing Company, 1979. 31. Salkever, op . ci ~ . 32. M. W. Jones-Lee. "The Value of Changes in ache Probability of Death or Injury, n Journal of Political. Economy, Vot. 82 ~ 1974), pp . 835 - 849 . 33. J. P. Acton. "The Value of Life: An Overview and Critique of Alternative Measures and Measurement Techniques, n Law and Contemporary Probl epic, Vat . 40 ( 1976 ), pp . 46 ~ 72 . . B. CO Conley. The Value of Hewn Life in the Demand for Safety, n American Economic Review, Sol. 66~1976), pp. 4S-55. 35 . W. B. Arthur. "The Economics of Risks to Life, n American Economic Review, Vot . 71~1981), pp . 54- 64 . 36. A. N. Bailey. "Research and Development Costs and Returns: lathe U. S . Pharmaceutical Industry, ~ Journal of Political Economy, tool. 80~1972), pp. 70-8S. . D. Schwartzman. The Expected Retune. from Pharmaceutical Research . Washington, DC: American Enterprise Institute for Public Policy Research, 1975 . - 909
